Unpacking Sam Altman’s latest statements on AGI.

Sam Altman, founder and CEO of OpenAI, has released a short blog again. He mainly focuses on three observations and more thoughts on AGI (Artificial General Intelligence). I will share his takes and add my own points of view on some of his statements and observations.

First, what is AGI? The term is vague (even for experts) and I personally don’t really know what it is that this will become. I think, in general, it stands for a form of super intelligent AI that can do anything a human can do, in any field, but better.

Altman’s blog starts with explaining the innovations that have brought humanity forward, mentioning that every time it improved the lives of people for the better. This is also a comment that I get most often when I try to explain that most people still underestimate the change this will bring. They will quickly say something like “we have seen big disruptive tech before and every time everyone was fine”. Although I agree that we will be fine, I do think this time it is different. Something Altman also mentions:

Why I believe it is different this time? Mainly speed and the fact we are talking about an actual new form of intelligence, possibly causing a very, very big change in the world.

His first observation is a technical one: “The intelligence of an AI model roughly equals the log of the resources used to train and run it.” I find it interesting that he states this, again, as there have been many rumors that there is a plateau reached when it comes to improving models by just throwing more data and compute at it. Initially, we of course saw DeepSeek shaking up the industry claiming frontier performance with a lot less compute.

In terms of speed, it is the second observations that Altman shares. I quote: “The cost to use a given level of AI falls about 10x every 12 months, and lower prices lead to much more use. You can see this in the token cost from GPT-4 in early 2023 to GPT-4o in mid-2024, where the price per token dropped about 150x in that time period. Moore’s law changed the world at 2x every 18 months; this is unbelievably stronger.”

That is insane. One of the things I call out most is that the speed of change will feel overwhelming (if it doesn’t already) and is something we have not witnessed before. To look at brands and corporates for example. We all know how long it took big brands to adopt internet, to adopt and adapt to social media. We are talking years and years. You might argue some brands didn’t really catch on at all. We are now expecting these same businesses to adapt to AI, AI Agents, and AGI within the next few years!

When the Fortune 500 list was first launched in 1955, the list contained mostly companies from manufacturing, mining, and energy sectors. Now, in its 2024 edition, only 49 of those survived. Even crazier, the current top-10 only holds 1 energy company (Saudi Aramco), 1 car manufacturer (Tesla) and 1 multi industry holding company (Berkshire Hathaway Inc by Warren Buffet). The others? Internet, Computers and Technology.

My question to you is, how do you think the Fortune 500 list will look like in the year 2035? The current technological evolution we are in is a big risk for companies, as they need to evolve to stay alive (in my humble opinion) and on the other hand it is a massive opportunity for new teams and brands.

What was the big difference between Amazon and maybe a large retailer trying the same? Amazon was internet native from the start, not only tech wise but also culture wise. The same will happen now, there are already brand new companies starting up that are fully AI native. They will be 10x faster to adopt, adapt and pivot then any other large organization. In my earlier newsletter I already mentioned it, the 1-person (AI native), 1 Billion USD company will happen.

That we have never seen this speed before is also subject of his third observation: “The socioeconomic value of linearly increasing intelligence is super-exponential in nature.” Meaning, that even small, steady improvements in intelligence can lead to huge leaps in value for society and the economy.

Another quote that I want to make a comment on is this: “The world will not change all at once; it never does. Life will go on mostly the same in the short run, and people in 2025 will mostly spend their time in the same way they did in 2024.”

This goes back to what I also mention often, maybe already in earlier editions. Most people overestimate the power of AI, and its changes, in the short run but UNDERestimate the impact on the longer run. So although you might think “everyone keeps talking about all the crazy stuff that will happen but nothing is changing” don’t let if fool you.

There are three last things that I want to talk about, coming from his article.

First, the importance of AGI benefiting all of humanity. It is challenging to talk about these matters as nobody can really see/understand yet in what shape or form “AGI” will happen. Will it be one specific entity/model that we talk to like GPT? Who will have access to it? Who decides what this AGI model needs to focus on? Or will there be millions of versions of it that we can all utilize? Again, it is very strange to think of this.

Then going towards the topic, how do we make sure everyone benefits? Altman wonders if every person on earth should get a personal “compute budget”, but also believes that making the intelligence so cheap will make it available for everyone anyway. He also hints to his, seems like, slowly changing believe on open sourcing AI. Currently, OpenAI’s models are closed and you can only access them via their own interfaces. On the other hand there are Meta’s Llama models, which are open source, and DeepSeek goes even further.

Should AGI, as powerful as people think it will be, be available to anyone or should it be controlled? It is a tough debate. In short, available for anyone means also people with bad intensions, but keeping it closed means it will be controlled by a handful of people. Both scary options in my opinion. Personally, I am more leaning towards open sourcing it and believing that “the good guys” will always build better tech that can help prevent “the bad guys” from doing damage with it. But it is also one of those topics that are hard to take a side on, so my thinking can definitely change on that in the future. The most important thing is making sure this technological change will benefit everyone and not only a few.

Second, he briefly mentions what skills and capabilities people should focus on in an age with more powerful AI. In short, he lists: agency, willfulness, determination, resilience and adaptability. I love, and agree, with every one of them. It is a question that gets asked often, both from a purely personal perspective but also from a school system perspective. What should we teach our kids? In my humble opinion, it is indeed all about soft skills. It will not work any longer to teach young adults specific frameworks, knowledge/facts, tools or processes as they will have changed, or become completely obsolete, a few months later. Courses and classes should probably be designed in a way that you can train (not learn) things like determination and agency.

All of it so you become, and stay, open and adaptable to any change that might happen, because that is the only thing we know for sure; sh*t will always keep changing (fast). If I may add to his list, I normally answer this question with two words: curiosity and humbleness. Genuine curiosity means you will dig into stuff that matters to you, without anyone telling you to do so, which keeps you developing with the changing times. Humbleness is important in order to always remember, you don’t know anything. Take nothing for granted and never feel “too good” to learn something new, because you will be left behind.

Third, last but not least, you might have wondered when I would say these words; AI Agents. YES! Haha, I told you before we will talk about AI Agents a lot this year and it seems like Sam Altman agrees with me 😉. I quote Altman: “Still, imagine it as a real-but-relatively-junior virtual coworker. Now imagine 1,000 of them. Or 1 million of them. Now imagine such agents in every field of knowledge work.” As mentioned before, there will be billions of agents. We will all become managers of some sort, managing our team of AI Agents helping us with both our work/business and life. The better the underlying AI models become (as discussed above) the better these agents will become. I will keep it short as I have talked about it many times before, but just to reaffirm you, this is happening.

We are living in an extraordinary time and we sometimes forget, because of all the bad news and real pain that is out there in the world, that there has been no better time to be alive then now. If you disagree with me, that is ok, but I do want to invite you to zoom out and look at how life was 100, or 500 years ago. I hope that will give you a more positive outlook on the world we have today.

It is also extraordinary because of what lies ahead as we are going to witness AGI, robotics, quantum computing, all in our lifetime. How crazy different will the world become? Let me round this up with some of Altman’s biggest statements in the article:

And:

Let those sink in for a bit. It is hard to grasp what this would really mean but yes it does get me hyped up! I am here for it. I believe we have to innovate and keep pushing forward, always keeping in mind the potential dangers. But not only dangers, also keeping in mind that it should benefit people and by that I mean THE people, all of us.

Let’s make sure this technology will be utilized to raise the bear minimum way of living on this planet, and not just for a few.

To end it, here is OpenAI’s Super Bowl ad for ChatGPT, showing human innovation at its finest.

Here are some of the latest announcements and other things that I found interesting this week (underscore means clickable!):

• To continue with OpenAI and Sam Altman. He just released an important tweet explaining what is to come in the (near) future from the company. Love it that they are simplifying things and making their strongest models available to anyone (including free subscriptions). Have a read:

• Replit’s AI Agent, who can build you software and apps by just using natural language, is now available on your phone. Develop prototypes on the go, the barrier to entry for development is being lowered once again.

• Elon Musk is leading a consortium of investors that announced it had submitted a bid of $97.4 Billion for “all assets” to OpenAI’s board of directors. Sam Altman quickly responded:

• Remember aixbt, the AI Agent on X that shares super-relevant crypto industry information? It currently has 484.000 followers, with 100.000+ people having their notifications on for its tweets. This is an autonomous AI Agent with real influence. Crazy, scary, exciting, all at the same time.

• Meta has achieved major breakthroughs in understanding human intelligence with AI models that decode brain signals to reconstruct typed sentences and map how thoughts become language. Crazy! Important, because how do we know what true artificial intelligence is, or when an AI is conscious, if we still don’t really understand our own intelligence and consciousness?

• Random big shoutout to my man Ragaco, former colleague, friend and AI Artist. Look at this image, completely generated. AI Generated Art, is still ART.

  

PS... If you’re enjoying my newsletter, will you take 6 seconds and refer this edition to a friend? It goes a long way in helping me grow the newsletter (and help more people understand our current technology shift). Much appreciated!

PS 2... and if you are really loving it and want to buy me some coffee to support. Feel free! 😉 

^{Who am I and why you should be here:}

^{Over the years, I’ve navigated industries like advertising, music, sports, and gaming, always chasing what’s next and figuring out how to make it work for brands, businesses, and myself. From strategizing for global companies to experimenting with the latest tech, I’ve been on a constant journey of learning and sharing.}

^{This newsletter is where I’ll bring all of that together—my raw thoughts, ideas, and emotions about AI, blockchain, gaming, Gen Z & Alpha, and life in general. No perfection, just me being as real as it gets.}

^{Every week (or whenever inspiration hits), I’ll share what’s on my mind: whether it’s deep dives into tech, rants about the state of the world, or random experiments that I got myself into. The goal? To keep it valuable, human, and worth your time.}